2009/11/1 Rob Browning <[email protected]>:
> Can you elaborate on the problem?

I'm attaching a Coq script that can be used to test the performance of
emacs23 with respect to emacs22.

Here are the basic instructions. You need at least Debian packages
coq, proofgeneral and proofgeneral-coq. On-site configuration includes
adding to ~/.emacs the line

(load-file "/usr/share/emacs/site-lisp/proofgeneral/generic/proof-site.el")

When proofgeneral is configured, you can open the attached Preorder.v
in emacs23 and emacs22, and compare performance by typing C-c C-b,
which makes proofgeneral communicate with the Coq interpreter (by
means of XML, I guess) and execute the entire script. On my laptop the
execution takes about 25 secs in emacs23 compared to just above 5 secs
in emacs22. In emacs23 you may see a surge in the CPU usage. On my
laptop this quickly causes the fan to throttle, etc. It's much more
convenient to use emacs22 because of this issue.

Also, I think I may have posted another report earlier that the
proofgeneral toolbar is empty in emacs23.

Best regards,
Vladimir

Attachment: Preorder.v
Description: Binary data

Reply via email to